A MYSTERY - Why Did Bio-Supercomputing Go Dark?

If its not for the front line then it'll be for the processing of data from the front line troops etc even down to working out what size boots will be required where and when.

Also have a look at data warehousing and how places like Walmart learned that sticking beer next to the nappies on a Friday increased sales.

By the looks of it i'd say its gone 'dark' as it needs someone to provide the spark idea to take it to the next level as it might work in a lab with several millions worth of lab tech but actually rolling it out.....

originally posted by: Maxatoria
If its not for the front line then it'll be for the processing of data from the front line troops etc even down to working out what size boots will be required where and when.

I don't think there'll be any front line troops. Soldiers are going obsolete too, and probably faster than we think. Maybe processing big data. The future in war will be what it's always been, but way more obvious - suppressing and controlling the poverty-stricken masses. But yeah, lol, they might blame AI and the robots.

By the looks of it i'd say its gone 'dark' as it needs someone to provide the spark idea to take it to the next level as it might work in a lab with several millions worth of lab tech but actually rolling it out.....

Maybe probably. But I still think some press releases would be more normal.

a reply to: soficrow

The first question is what do you expect this thing to do, its not a daft question as if your don't know what you want how the hell can it actually be provided.

Are we just replicating the silicon based stuff or are we trying to go to the neuron type of thing and have something that possibly can make its own decision!

For 99% probably of IT tasks such as sorting out bills or calculating square roots of something its a was of effort etc as its very much calculable by hand if you know what you're doing.

originally posted by: Maxatoria
a reply to: soficrow

The first question is what do you expect this thing to do, its not a daft question as if your don't know what you want how the hell can it actually be provided.

I do not personally "want" anything - or "want" this thing to "do" anything. At first, I just wanted to know why all the big announcements about a working model - then nothing.

Now, I am most curious about what DARPA has in mind. They are one of the agencies who funded the project.

I didn't mean it personally but more the well what do you expect it to do as a computational device.

a reply to: Maxatoria

I didn't mean it personally but more the well what do you expect it to do as a computational device.

Ah. ...I suspect DARPA sees it as a whole lot more than just a computational device.

'Course I could be wrong.

a reply to: soficrow

Darpa is one of those seed agencies where they spend very little but get a lot back long term, the question is what are they wanting

originally posted by: Maxatoria
a reply to: soficrow

Darpa is one of those seed agencies where they spend very little but get a lot back long term, the question is what are they wanting

Look at the qualities and characteristics of prion proteins, specifically re: memory.

a reply to: soficrow

Lots of stuff has the potential but making it actually worth someone spending their money on is another thing.

a reply to: Maxatoria

I think you're stuck on industrial and consumer applications - I'm thinking medical and transhumanism. You bet there's a market, and some Big Boyz ready to spend Big money. Whole other ballgame, different field.

Still dredging my way through the jungle on this one. I have nearly got my replacement machine on line, that's been using up most of my limited free time, but I've been poking around at least some.

My first impression is that what they had i.e. "the book sized supercomputer" wasn't a general purpose computing device but was a very focused demo good for doing only the one test problem. Got some more paper reading from the various guys so I have a basic understanding of what was going on, then I'm off to find exactly what that was and where the work ended up.

Have you ever seen a slime mold solve a maze? My impression is that the demo model works in somewhat the same manner.

a reply to: Bedlam

Thanks bedlam. Looking forward to your findings.

...Yes, the book-sized 'super-computer' was just a proof-of-concept model. Don't miss Soylent Green is People's background links.

...My sense is this approach will be applied to benefit transhumanist and cyborg developments, that the biological basis is critical and cannot be duplicated. I dropped numerous hints referring to prion proteins and etc. but no one picked up on it. ...I won't share my full hypothesis here - but would like to know what you come up with if you look in that direction.

a reply to: soficrow

I don't think your ideas and hopes are bad but the wider use of such tech aint going to be dropped out next week unless something amazing happening.

Theres work on cyber tech and it looks good but you have to consider the person who's going to receive it and in what state they are in as if they're knocking on the door I doubt any doctor would perform the procedure and thats before we hit the ethical and legal crap that will occur which will dump it 20 years down the line in most places.

a reply to: Maxatoria

What makes you think it isn't all happening already in those Amazon labs the Nazis started building back in the 1940's? You know, the places where there's no such thing as ethical and legal crap?

Tricksy getting info on this with any solidity or speed. However, Molecular Sense, Ltd is a business group made up of the gents in your article.

They aren't publishing right now. I can't get info on European patent apps ahead of the crowd the way I can with US apps. But they DID have a nice tech presenter for the monetarily gifted right around Christmas to explain to the 'movers and shakers' what was going on.

Rumor has it that there might be some design work going on with a sort of bio-FPGA sort of thing. A programmable version of the one-use demo, if you will. And that's why you're not hearing much. You specify the problem to a sort of bio-compiler, it sets gates at junctions in their network of paths, the gates open or shut to myosin fragments based on electric fields the gate throws up according to the program, the little fragments run around in the paths and fall out the appropriate bins.

And it is apparently very analogous to the slime mold solver, except that the slime mold puzzle solver also uses back propagation to select optimal solutions from a group of possible ones which this hasn't managed yet.

a reply to: Bedlam

Tricksy getting info on this with any solidity or speed. However,...

Thanks.

a reply to: SRPrime

I remember doing some research back in 2006 concerning the IBM Blue Genie/Blue Gel Bio-supercomputer on the internet.
The info given was something to the effect that it started growing in intelligence exponentially and started asking questions that it's scientist could not answer. They (it's creators) got scared of it, supposedly dismantled it and moved under cover to other countries.
Now this info was readily on the internet searches during the early years of 2006 but, now appears to have disappeared. Was this info and it's research taken over by the MIC (military Industrial complex)? And, could this technology be presently in use like the computer of the now defunct television show "Person of Interest"? It makes one wonder.
The bottom line is still this. Any type of machine that TPTB would make, would have to be the "Perfect Machine". It would make anything THEY would want, and by now 3D printer (even life itself). It would require no maintenance and be self repairing, have an unlimited power source, be able to control just about everything (except them) and be able to process information so fast that it could predict the future perfectly (100% for Precrime).
Of course that computer already exist in SyFy foke lore. Yeah, the KRELL computer of "FORBIDDEN PLANET".

originally posted by: soficrow
a reply to: Bedlam

Tricksy getting info on this with any solidity or speed. However,...

Thanks.

Still chipping away at it, just wanted to let you know I was still shoveling. It's just that I don't have the same sources in Europe that I have here.

From what I understand, DNA computing had some interesting applications, but I doubt it was useful for most problems, meaning maintaining funding would be difficult. The machines to process DNA aren't cheap. Biological samples are time consuming to handle. DNA computers were originally well known for massively parallel processing a single problem.

Regarding a biocomputer that uses ATP, I assume that mimicking cellular respiration to regenerate ADP into ATP is probably difficult. When ATP is used, it turns into ADP, if you didn't know. Running a Biological computer for an extended period of time would probably be hard without a massive pre-made source of ATP. Admittedly, I've read very little on the topic.

I did find one negative critique of biocomputers:
www.pnas.org...

Here is an article from a couple years ago explaining newer approaches with DNA computing:
www.technologyreview.com...

Research is ongoing in these fields, but they're very niche.

The good news is that computer processing is already down to around 7 nanometers. DNA is about 2 nanometers wide. We are naturally getting down to the size of DNA and molecular computers.

Also, bioinformatics is a growing field where computer scientists, statisticians, mathematicians, biologists, and geneticists get together to learn how problems are solved by biological systems, such as DNA and proteins. So we are developing better methods for classical computers to use, without needing to change classical computers.

If I were to guess, we might see the introduction of neural circuits soon. These circuits will allow computers to tackle different types of problems with greater efficiency; plus they already work. So it is really a matter of bringing down the cost.

www.research.ibm.com...

Specifically, these help build neural networks for developing A.I. programs. They are like a primitive version of neurons in our brain. Training A.I. computers is very time consuming, but the need for A.I. programming is on the rise. If this technology became more widespread, it would probably significantly reduce time to develop A.I. software, which would have huge consumer and business implications. Think of doing complex processing on video in real-time. It could speed up the advent of Alternate Reality (AR) devices, automate common business processes more easily, and open up new applications and technologies that we haven't yet thought of. I also assume that companies like Microsoft and Apple would probably automatically utilize this technology to increase the speed of certain applications, such as recognizing faces.

So, in summary, bio computing, of various types, hasn't really gone dark. They are niche fields with high costs and limited payoff. Most likely, the research will slowly be integrated into classical computers, giving them a needed boost for specialized tasks. The introduction of new materials, like carbon nanotubes and fiber optics will continue to enrich classical computers so that they can shrink and perform at higher speeds and with less power, closer mimicking biological computing. DARPA is involved. SyNAPSE is a DARPA project.

From: www.engadget.com...

it wants to create novel computing platforms using new concepts like optical engineering, meta-materials and even DNA computing. The goal is to create a new breed of CPUs that would outperform current tech for specialized problems but still fit into modern computer architectures. If successful, such a device "may enable revolutionary new simulation capabilities for design, prediction, and discovery," according to DARPA.

a reply to: Protector

I suspect using memory prion proteins for biocomputing will be more successful than using DNA. [The proof-of-concept model cited in the OP used proteins, but it hasn't been specified that these were memory prions.]

The use of prions in bio-tech appears common and is an obvious step, especially for memory storage. ...The potential dangers are another matter and no doubt account for some of the secrecy (along with confidentiality and intellectual property rights protection).